Sunday, Jan. 3, 2021 10:00 AM - 12:00 PM (EST)
- Chair: Graciela Laura Kaminsky, George Washington University and NBER
Origins of Serial Sovereign Default
AbstractWhy do countries become serial sovereign defaulters? Using the full history of external sovereign debt issuance and default from 1820 to the present, we explore how the political and economic environment in which countries borrow and default affects their likelihood of become a 'serial defaulter'. We categorize countries as 'serial defaulters' using several criteria based on the frequency, duration, and scale of default. We collect new quantitative and narrative evidence on countries' reasons for initial debt issuance and causes of default. We calculate how a countries' likelihood of both becoming a serial defaulter and re-entering external debt markets varies under different borrowing and default conditions. We aim to investigate how adverse initial conditions such as wars, shifts in political power, the nature of economic crises, and underwriter market power influence who becomes a serial defaulter above and beyond economic fundamentals.
Two Hundred Years of Rare Disasters: Financial Center Crises, Social Unrest, and Pandemics
AbstractThe 2007-2009 US Subprime Crisis was not the usual type of crisis. This crisis had the financial center at the epicenter, with international liquidity collapsing and the crisis spreading globally. The world is now immersed in a new crisis. This time around, countries around the world are fighting the spread of a global pandemic. These events are rare disasters, only erupting on average every 50 or 100 years. Do all rare disasters have similar adverse and protracted effects on the economy? What are the effects on financial markets? Do financial markets anticipate the extent and persistence of the collapse of the economy?
To study these rare disasters, we construct a database on sovereign bond yields and economic activity for all countries borrowing in international capital markets starting in 1820. We study eight events, five financial crises with the financial center at its epicenter (the London Panic starting in 1825, the 1873 Financial Crisis starting in Germany and Austria, the Barings Crisis in 1890, the Great Depression starting in 1929, and the US Subprime Crisis starting in 2007), two pandemics (the Spanish Flu starting in 1918, and the Coronavirus Pandemic starting in 2019), and one episode of social revolts (the Revolutions in 1848-1849). Our results indicate that crises during the first episode of financial globalization (starting at the end of the Napoleonic Wars and ending with the Great Depression) tend to have quite persistent effects both on financial markets and on economic activity. In contrast, during the second episode of financial globalization (starting with the collapse of the Bretton Woods System) there is a disconnect between financial markets crashes and economic activity. While downturns in economic activity following a crisis in the financial center continue to be quite protracted, the effects on financial markets are far less persistent.
Answering the Queen: Machine Learning and Financial Crises
AbstractFinancial crises cause economic, social and political havoc. We use the general framework of sequential predictions also called online machine learning to forecast crises out-of-sample. Our methodology is based on model averaging and is meta-statistic since we can incorporate any predictive model of crises in our set of experts and test its ability to add information. We are able to predict systemic financial crises 12 quarters ahead in quasi-real time with very high signal to noise ratio. We analyze which experts provide the most information for our predictions at each point in time, allowing us to gain some insights into economic mechanisms underlying the building of risk in economies. Finally, we show that our methodology is also able to predict systemic crises in France using real time data.
- F3 - International Finance
- F4 - Macroeconomic Aspects of International Trade and Finance